Machine Learning

Dr. Rolando A. Pula · 4.7 (48,911) Bestseller

ai machine_learning 1hr

This course provides a practical introduction to the fundamental principles, algorithms, and real-world applications of Machine Learning (ML). Designed for aspiring data scientists, AI developers, and software engineers, this program bridges theoretical mathematical foundations with modern hands-on implementation.

Students will cover the entire machine learning pipeline—from exploratory data analysis and feature engineering to model training, hyperparameter tuning, evaluation, and deployment. By the end of this course, learners will be capable of identifying appropriate ML algorithms for complex real-world problems and building end-to-end predictive systems.

Approved by Partners

Legit and professional course

This course is backed by trusted partner organizations, highlighting strong industry alignment, recognized quality, and professional credibility.

PEZA

Certified Track

DOST

Certified Track

University of Perpetual Help

Certified Track

NEXUSPRO

AI · ML · DL

A visual guide to how machines get smart

ARTIFICIAL
INTELLIGENCE

...and the two ideas nested inside it that make modern AI work: machine learning, and deep learning.

The hierarchy

One field, nested three layers deep

Each layer sits fully inside the one before it.

Artificial Intelligence

The broad field: any system built to perform tasks tied to human intelligence.

Machine Learning

AI that learns its own rules by finding patterns in data.

Deep Learning

Machine learning built from many stacked neural-network layers.

Layer 01 · Outer ring

Artificial Intelligence

The broad ambition of the field: building systems capable of performing tasks we normally associate with human intelligence — however that capability is achieved.

Speech Recognition

Turning spoken audio into text and intent.

Language Understanding

Parsing meaning, context, and intent from text.

Planning & Decisions

Weighing branching options to choose an action.

Robotics & Control

Sensing an environment and acting within it.

Layer 02 · Middle ring

Machine Learning

Instead of hand-coding every rule, a machine-learning system is shown many examples — tables, images, sensor readings, labeled cases — and learns the underlying pattern itself.

MODEL

finds the pattern

Structured examples flow into a model, which outputs a prediction or classification.

LEARN FROM DATA

Layer 03 · Inner core

Deep Learning

A specific way of doing machine learning: stack many layers of artificial neurons, and let the network build up its own increasingly complex features.

Data enters on the left and passes through hidden layers to reach an output on the right.

MULTIPLE NEURAL NETWORK LAYERS

Side by side

Same hierarchy, three lenses

Artificial Intelligence

The broad field of building systems that perform tasks associated with human intelligence.

e.g. a chess engine following hand-written rules is still AI — with no learning involved.

Machine Learning

AI systems that learn their own patterns from data instead of following fixed, hand-written rules.

e.g. a spam filter trained on thousands of labeled emails.

Deep Learning

Machine learning that uses artificial neural networks with many stacked layers.

e.g. an image classifier built from dozens of network layers.

01 / 07

What you'll learn

This course works through thirteen classical models, grouped the way they're taught: first by how the model learns, then by the task it solves. Each one is paired with the scikit-learn estimator you'll use in the labs.

Regression 2 Classification 5 Ensemble 3 Clustering 2 Dimensionality reduction 1

Supervised learning

The model trains on labelled examples — inputs paired with known, correct answers — and learns to predict the answer for new, unseen inputs. Three tasks live here: predicting numbers, predicting categories, and combining many models into one.

Regression

Predicting a continuous, numeric value

Linear Regression

linear_model.LinearRegression

Predicts a continuous number by fitting the straight line that best relates the input variables to the output.

Applications: House price forecasting, demand estimation, dose–response curves.

Regularised Regression (Ridge & Lasso)

linear_model.Ridge · linear_model.Lasso

Improves plain linear regression by penalising large coefficients, which prevents overfitting and can drop weaker features entirely.

Applications: Wide datasets with many columns, automatic feature selection, genomics.

Classification

Sorting data points into labelled categories

Logistic Regression

linear_model.LogisticRegression

Predicts the probability of a categorical outcome, like yes or no, using an S-shaped curve instead of a straight line.

Applications: Customer churn, credit scoring, medical risk screening.

Decision Tree

tree.DecisionTreeClassifier

Splits data into branches based on feature values, ending in leaves that each represent a final predicted class.

Applications: Loan approval rules, clinical triage, any model that must be explained.

K-Nearest Neighbors (KNN)

neighbors.KNeighborsClassifier

Classifies a new point by taking a majority vote among its closest neighbours already present in the data.

Applications: Recommendation engines, handwriting recognition, anomaly spotting.

Naive Bayes

naive_bayes.GaussianNB · MultinomialNB

Estimates the probability of each class using Bayes' theorem, assuming the input features are independent of one another.

Applications: Spam filtering, sentiment analysis, document tagging.

Support Vector Machine (SVM)

svm.SVC · svm.LinearSVC

Finds the boundary that separates two classes by the widest possible margin between their closest points.

Applications: Text categorisation, image classification, bioinformatics.

Ensemble methods

Combining many simple models into one strong one

Random Forest

ensemble.RandomForestClassifier

Trains many decision trees on random subsets of data and averages their votes for a more accurate, stable prediction.

Applications: Fraud detection, credit risk, feature-importance analysis.

Gradient Boosting

ensemble.HistGradientBoostingClassifier

Builds an ensemble of weak models one at a time, where each new model corrects the errors left by the ones before it.

Applications: Search ranking, click-through prediction, tabular benchmarks.

AdaBoost

ensemble.AdaBoostClassifier

Trains a sequence of simple classifiers, giving more weight each round to the examples the previous ones got wrong.

Applications: Face detection, quick screening models, small tabular datasets.

Unsupervised learning

These models never see a labelled answer. They explore the raw structure of the data itself — grouping similar points together, or compressing many features down to the few that carry the most information.

Clustering

Grouping unlabelled points by similarity

K-Means Clustering

cluster.KMeans

Groups unlabelled data into a set number of clusters by repeatedly assigning each point to its nearest cluster centre.

Applications: Customer segmentation, image compression, document grouping.

Hierarchical Clustering

cluster.AgglomerativeClustering

Builds a tree of nested clusters by progressively merging the most similar groups of data points together.

Applications: Taxonomy building, gene expression grouping, org-structure analysis.

Dimensionality reduction

Compressing many features into the few that matter

Principal Component Analysis (PCA)

decomposition.PCA

Reduces the number of features in a dataset while preserving as much of its original variance as possible.

Applications: Visualising high-dimensional data, noise reduction, speeding up training.

Quick reference

Start from the question you're trying to answer, then read across.

If your question is…	Task	Start with	Labels needed
“How much?” or “How many?”	Regression	`LinearRegression`, `Ridge`	Yes
“Which category does this belong to?”	Classification	`LogisticRegression`, `SVC`	Yes
“Can I squeeze out more accuracy?”	Ensemble	`RandomForestClassifier`	Yes
“What natural groups exist here?”	Clustering	`KMeans`	No
“Too many columns to work with.”	Dim. reduction	`PCA`	No

Explore related topics: ai machine_learning

This course includes:

1hr of video lessons with self-paced lab works
Certificate of completion

Teaching Machines to Learn: Your Journey into Artificial Intelligence

Course overview

Course roadmap

Select a lesson from the right panel to start guided learning.

Module 1: Hands-on Machine Learning

8 items

Buy this course to unlock all lessons and advanced tools.

Free

30-day money-back guarantee

Certificate of completion

Unlockables

CoLab badges

All badges

Complete course evidence to unlock these credentials.

Machine Learning Explorer

Explorer · ML-EXP

Locked

Machine Learning

Approved by Partners

ARTIFICIALINTELLIGENCE

Supervised learning

Regression

Linear Regression

Regularised Regression (Ridge & Lasso)

Classification

Logistic Regression

Decision Tree

K-Nearest Neighbors (KNN)

Naive Bayes

Support Vector Machine (SVM)

Ensemble methods

Random Forest

Gradient Boosting

AdaBoost

Unsupervised learning

Clustering

K-Means Clustering

Hierarchical Clustering

Dimensionality reduction

Principal Component Analysis (PCA)

Quick reference

Teaching Machines to Learn: Your Journey into Artificial Intelligence

Course roadmap

Module 1 Hands-on Machine Learning 8 lessons

Module 1: Hands-on Machine Learning

Select a lesson

ARTIFICIAL
INTELLIGENCE